Research map / Adversarial robustness and security / Attacks on text and graphs
Attack Can Benefit: An Adversarial Approach to Recognizing Facial Expressions under Noisy Annotations
Jiawen Zheng, Bo Li, Shengchuan Zhang, Shuang Wu, Liujuan Cao, Shouhong Ding
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Apply GAAVE, Dataset Splitting and Subset Refactoring, and Self-Annotator | ResNet-18 baseline · RAF-DB | Acc. (%): 86.93 -> 89.29 | Improved |
| Apply GAAVE, Dataset Splitting and Subset Refactoring, and Self-Annotator | ResNet-18 baseline · FERPlus | Acc. (%): 88.20 -> 89.83 | Improved |
| Apply GAAVE, Dataset Splitting and Subset Refactoring, and Self-Annotator | ResNet-18 baseline · AffectNet-7 | Acc. (%): 60.34 -> 65.60 | Improved |
| Apply GAAVE, Dataset Splitting and Subset Refactoring, and Self-Annotator | ResNet-18 baseline · AffectNet-8 | Acc. (%): 59.38 -> 62.78 | Improved |
| Add GAAVE (ablation) | ResNet-18 FER training · RAF-DB | Acc. (%): 86.93 -> 88.04 | Improved |
| Add Dataset Splitting and Subset Refactoring (ablation) | ResNet-18 FER training · RAF-DB | Acc. (%): 86.93 -> 87.79 | Improved |
| Add Self-Annotator alongside GAAVE (ablation) | ResNet-18 FER training without Dataset Splitting and Subset Refactoring · RAF-DB | Acc. (%): 88.04 -> 88.46 | Improved |
| Add Self-Annotator alongside GAAVE and Dataset Splitting and Subset Refactoring (ablation) | ResNet-18 FER training · RAF-DB | Acc. (%): 88.85 -> 89.29 | Improved |
| Add GAAVE (ablation) | ResNet-18 FER training · AffectNet-7 | Acc. (%): 60.34 -> 63.37 | Improved |
| Add Dataset Splitting and Subset Refactoring (ablation) | ResNet-18 FER training · AffectNet-7 | Acc. (%): 60.34 -> 61.92 | Improved |
| Add Self-Annotator alongside GAAVE (ablation) | ResNet-18 FER training without Dataset Splitting and Subset Refactoring · AffectNet-7 | Acc. (%): 63.37 -> 64.37 | Improved |
| Add Self-Annotator alongside GAAVE and Dataset Splitting and Subset Refactoring (ablation) | ResNet-18 FER training · AffectNet-7 | Acc. (%): 63.93 -> 65.60 | Improved |
Closest papers by meaning
- Navigating Label Ambiguity for Facial Expression Recognition in the Wild (AAAI 2025 · similarity 0.89)
- Cross-Domain Facial Expression Recognition: A Unified Evaluation Benchmark and Adversarial Graph Learning (TPAMI 2021 · similarity 0.88)
- LA-Net: Landmark-Aware Learning for Reliable Facial Expression Recognition under Label Noise (ICCV 2023 · similarity 0.87)
- Learn from All: Erasing Attention Consistency for Noisy Label Facial Expression Recognition (ECCV 2022 · similarity 0.87)
- Leave No Stone Unturned: Mine Extra Knowledge for Imbalanced Facial Expression Recognition (NeurIPS 2023 · similarity 0.87)
- Open-Set Facial Expression Recognition (AAAI 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
